Authors' objectives

To evaluate the effectiveness of bibliotherapy for sexual dysfunction, and to examine correlations in treatment effects.

Searching

The following sources were searched for both published and unpublished studies: MEDLINE from 1966 to 1996; PsycLIT from 1974 to 1996; chapters and books from 1987 to 1996; and references from relevant journal articles and books. The search words were 'bibliography', 'self-help', 'self-help techniques', 'sexual dysfunctions', 'sexual disturbances', 'sexual problems' and 'sexual disorders'.

Study selection

Assessment of study quality

The studies were coded according to the following.

1. Whether there was a comparison group with a waiting-list, no treatment, or placebo control.

2. Whether the allocation of the participants was random, non-random, or unspecified.

3. Whether the drop-out rate was low (less than or equal to 25%), high (greater than 25%), or unspecified.

The internal validity was classified as either low, medium, or high depending upon the extent to which the above criteria were met.The author stated a method for assessing the validity, but did not state how many of the reviewers performed the assessment.

Data extraction

The author extracted and coded the following data:

the type of sexual dysfunction;

the method used to administer bibliotherapy;

the medium used in bibliotherapy;

the duration of the treatment (less than or equal to 10 weeks, or greater than 10 weeks);

the characteristics of the study sample (volunteers, patients, or mixed sample);

the compliance rate (less than or equal to 75%, greater than 75%, or unspecified);

the comparison group (waiting-list, no treatment, placebo, or other treatment);

the method used to for participant allocation (random, non-random, or unspecified); and

the drop-out ratio, classified as low (less than or equal to 25%), high (greater than 25%), or unspecified. The effect size (ES) was calculated for every dependent measure in a study using the method of Glass et al. (see Other Publications of Related Interest no.1).

When the mean score and standard deviations of the control group were not reported, the ES was calculated using the method of Glass et al. (see Other Publications of Related Interest no.1). Non significant results were assumed to have a P-value of 0.5, unless otherwise specified, and their corresponding ES was set to 0. For dichotomous outcome variables, the ES was obtained through probit transformation (see Other Publications of Related Interest no.1). The dependent variables that were listed in the methods sections, but not reported in the results section, were assumed to be non significant (P=0.5, ES=0). The mean ES for each comparison was obtained by calculating the arithmetic mean of the ESs of all the dependent variables.

Methods of synthesis

Results of the review

Twelve studies (397 patients) from 16 bibliotherapy groups were included in the meta-analysis. Nine of the 12 included studies were RCTs.

Validity: 19 studies were rated as having low internal validity, 56 as having medium internal validity, and 25 as having high internal validity.

The unweighted MES was 0.68 (95% CI: 0.23, 1.14) at post-treatment and 0.30 (95% CI: -0.10, +0.72) at follow-up. The unweighted ESs were inversely correlated with sample size, but this was not statistically significant (post-treatment, P=0.08; follow-up, P=0.27).

The weighted MES was 0.50 (95% CI: 0.27, 0.72) at post-treatment and 0.21 (95% CI: -0.08, +0.50) at follow-up. The corresponding weighted ESs ranged from -0.28 to +1.51, and from -0.44 to +0.79, respectively. The overall MES weighted for sample size Z was 4.07 (P<0.001) at post-treatment. A one-sample t-test showed that the observed MES at post-treatment (0.50) was significantly different from zero (t=4.74, d.f.=15, P=0; 95% CI: 0.27, 0.72).

The estimated fail-safe N of the weighted MES was 86 (proposed tolerance level was 90; see Other Publications of Related Interest no.2).

MES at follow-up (9 studies).

There was erosion of treatment gains at follow-up with no significant effect of treatment (Z=1.37, P=0.85); the weighted MES was 0.21 (95% CI: -0.08, +0.50; one-sample t=1.69, d.f.=8, P=0.129).

Authors' conclusions

The efficacy of bibliotherapy has, as yet, not been evaluated sufficiently for other sexual dysfunctions or in comparison with other therapeutic approaches for sexual dysfunction.

CRD commentary

The aims were stated and the inclusion criteria were defined in terms of intervention and study design, although the participants and outcomes were not defined a priori. The literature search included two relevant databases and a search for both published and unpublished material. Publication bias was assessed using the fail-safe N value. It was not reported whether any restrictions on publication language were applied. The author used defined criteria to assess validity. Some relevant information on the individual studies was presented in tabular format, although no details were given of the methods used to assess the outcomes in the individual studies. Data were pooled in a meta-analysis without a prior assessment of statistical heterogeneity, and the influence of study validity on the results was not examined. The discussion considered some of the limitations of the review. The suitability of outcome ascertainment methods cannot be assessed without details of the methods used to assess outcomes in individual studies. In addition, the appropriateness of the meta-analysis cannot be judged without an evaluation of statistical heterogeneity. In view of these problems, caution should be applied when considering the results.

Implications of the review for practice and research

Practice: The author states that the available evidence warrants the recommended use of self-help books for sexual dysfunction, but only after proper assessment.

Research: The author did not state any implications for future research.

Funding

The Netherlands National Foundation of Mental Health, grant number 4142.

Bibliographic details

van Lankveld J J. Bibliotherapy in the treatment of sexual dysfunctions: a meta-analysis. Journal of Consulting and Clinical Psychology 1998; 66(4): 702-708. [PubMed: 9735590]

Other publications of related interest

1. Glass GV, McGraw B, Smith ML. Meta-analysis in social research. Beverly Hills (CA): Sage Publications; 1981. 2. Rosenthal R. Meta-analytic procedures for social research. Newbury Park (CA): Sage Publications; 1991.

Indexing Status

Subject indexing assigned by NLM

MeSH

Bibliotherapy /standards; Clinical Trials as Topic /statistics & numerical data; Confidence Intervals; Female; Humans; Male; Sexual Dysfunctions, Psychological /therapy; Treatment Outcome

AccessionNumber

11999003058

Database entry date

31/05/2002

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.